Fostering Energy-Awareness in Simulations behind Scientific Workflow Management Systems
Fostering Energy-Awareness in Simulations behind Scientific Workflow Management Systems
复制标题
在科学工作流程管理系统背后的模拟中培养能源意识
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
R. Prodan
中科院分区：
文献类型：
--
作者：
G. Kecskeméti;S. Ostermann;R. Prodan
科学的工作流管理系统在云计算时代面临着新的挑战。过去关于所用基础设施状态的丰富信息的可用性已经消失。因此，组织虚拟基础设施，使其不仅支持正在执行的工作流，而且优化多个服务级别目标（例如，最大能源消耗限制、成本、可靠性、可用性）变得依赖于良好的基础设施建模和预测技术。虽然模拟器在过去已经成功地用于辅助对这样的工作流管理系统的研究，但是当前可用的云相关模拟工具包存在若干问题（例如，可扩展性，范围狭窄），这阻碍了它们的适用性。为了解决这一需求，本文介绍了技术，统一现有的两个仿真工具包，首先分析了当前的模拟器的问题，然后通过说明工作流系统所面临的问题，通过ASKALON环境的例子。最后，我们将展示如何统一的选定的模拟器改善所讨论的问题。
Scientific workflow management systems face a new challenge in the era of cloud computing. The past availability of rich information regarding the state of the used infrastructures is gone. Thus, organising virtual infrastructures so that they not only support the workflow being executed, but also optimise for several service level objectives (e.g., Maximum energy consumption limit, cost, reliability, availability) become dependent on good infrastructure modelling and prediction techniques. While simulators have been successfully used in the past to aid research on such workflow management systems, the currently available cloud related simulation toolkits suffer form several issues (e.g., Scalability, narrow scope) that hinder their applicability. To address this need, this paper introduces techniques for unifying two existing simulation toolkits by first analysing the problems with the current simulators, and then by illustrating the problems faced by workflow systems through the example of the ASKALON environment. Finally, we show how the unification of the selected simulators improve on the the discussed problems.